In the first quarter of 2026, Baidu's general business revenue reached RMB 26 billion, with AI-driven new business revenue contributing RMB 13.6 billion, accounting for 52%. Revenue from intelligent cloud infrastructure surged by 79% year-on-year, while GPU cloud revenue soared by 184%.

Robin Li made it clear this time: AI has become Baidu's core growth driver.

The key takeaway from Baidu's Q1 earnings report is not the total revenue of RMB 32.1 billion or single-quarter profit fluctuations, but rather a more fundamental structural change: AI revenue has, for the first time, accounted for more than half of Baidu's general business revenue.

I believe this signal is crucial for Baidu, but more importantly, it may represent a new turning point for China's internet industry.

Over the past two years, nearly all internet companies have been talking about AI. They have launched large models, upgraded assistants, introduced intelligent agents, integrated AI into office workflows, transformed search engines, and revamped customer service. While these developments are exciting, what the capital market truly wants to see is straightforward: Has AI made it onto the revenue statement? Has it become a core part of the main business? Has it altered the company's valuation anchor?

Baidu's answer this time is relatively clear: AI is no longer just a strategic slogan but has begun to transform the revenue structure itself.

I. Baidu Has Truly Changed: The Market Can No Longer Price It Solely as a 'Search Advertising Company'

In the past, many investors viewed Baidu primarily as a search advertising company.

This label is deeply ingrained. Users search for keywords, advertisers buy traffic, and the platform sells ad placements, allowing Baidu to earn high-margin advertising revenue. This model was once powerful and stable, but in recent years, the market has increasingly priced in risks. Traditional advertising budgets are susceptible to macroeconomic influences, and search traffic is being diverted by short-video platforms, content platforms, and app ecosystems. As long as the market views Baidu as a search advertising company, its valuation will struggle to truly break out.

Therefore, the most critical aspect of this Q1 earnings report is not whether Baidu's advertising business faces short-term pressure but rather that Baidu has presented an alternative growth anchor.

AI-driven new business revenue reached RMB 13.6 billion, accounting for 52% of general business revenue. This proportion carries significant weight. It means AI is no longer a peripheral business or a story that 'might materialize in three years' but has begun to occupy a central position in Baidu's revenue statement.

I see a substantial expectation gap here: The market used to ask whether Baidu's search advertising business could return to its former glory. Now, the more pertinent question is whether AI can become the new pricing logic for Baidu's core business.

These two questions point in entirely different directions.

If we focus solely on search advertising, Baidu is waiting for a recovery in an old track ( track : industry track). If we look at AI revenue accounting for over half, Baidu has already entered the realization phase in a new track . The former corresponds to valuation suppression, while the latter corresponds to valuation switching.

The pricing logic for Chinese internet companies in the past revolved around traffic, advertising, gaming, e-commerce commissions, and local services fulfillment efficiency. Companies with more users, higher transaction frequencies, and more valuable ad inventory enjoyed valuation flexibility. However, in the AI era, the market will focus more on another set of metrics: whether model capabilities have been commercialized, whether cloud revenue has scaled, whether intelligent agents have been integrated into business processes, whether AI applications have altered conversion rates and user engagement, and whether capital expenditures have translated into revenue growth.

Baidu is at the forefront of this transition. It has a search entry point, AI cloud capabilities, the Wenxin large model, an intelligent agent ecosystem, and physical-world scenarios like Apollo Go ( Apollo Go ). Among Chinese internet companies, Baidu is one of the few that can connect AI from models and computing power to applications and real-world scenarios.

This is why I believe this earnings report marks an industry turning point: Baidu is not just transforming itself but also providing a blueprint for Chinese internet companies—AI revaluation must ultimately be reflected in the revenue structure.

II. AI Cloud, AI Applications, and Robotaxi: Baidu's Capital Narrative Now Has Three Layers

Baidu's AI narrative is more comprehensive than simply 'building large models.'

The first layer is AI cloud.

In Q1, Baidu's intelligent cloud infrastructure revenue reached RMB 8.8 billion, up 79% year-on-year, while GPU cloud revenue surged by 184%. These figures send a clear message to the capital market: Enterprise demand for AI computing power, training, inference, and model deployment continues to rise, and Baidu Cloud is capturing this industrial upswing.

AI investments have often been viewed by the market as cost pressures. Purchasing servers, stockpiling GPUs, and maintaining R&D teams all weigh on short-term profits. When internet companies talk about AI, investors are excited but also concerned about deteriorating cash flow quality. However, with the high growth of AI cloud revenue, the nature of capital expenditures has changed. They are no longer just expenses but have the potential to translate into order visibility and revenue flexibility.

The second layer is AI applications, particularly the transformation of search and marketing.

Baidu App's MAU remains above 600 million, representing a massive information entry point. For Baidu, AI is not just about making the search bar smarter but about transforming the advertising transaction logic.

Traditionally, search advertising sold keywords and ad placements, with advertisers buying impressions and clicks. If AI-powered search can better understand user intent and deliver more comprehensive results, advertising could shift from 'buying traffic' to 'buying conversions, services, and outcomes.' This is crucial for Baidu because it is not merely patching up the old advertising model but replacing its underlying foundation.

In my view, the value of Baidu's AI-native marketing lies not just in revenue growth but in its potential to reimagine search monetization. Search advertising, once seen as a mature business, could become a new entry point for discussion thanks to AI-powered search.

The third layer is Robotaxi, represented by Apollo Go ( Apollo Go ).

This segment may not contribute the largest profits in the short term, but it is vital to Baidu's valuation narrative. It brings Baidu's AI from the online world to the offline, from the information realm to the physical world. The sustained growth in autonomous driving operation orders and the expanding cumulative order volume demonstrate that Baidu is not just experimenting with AI in labs but deploying it in real-world traffic scenarios. This sets Baidu apart from many other internet companies.

Alibaba's AI focus is on e-commerce and cloud, Tencent's on WeChat, gaming, and advertising, and JD.com's on supply chain and AI hardware. Baidu has added a physical-world dimension. AI cloud drives revenue flexibility, AI search and marketing transform the core business, and Robotaxi opens up long-term valuation potential.

The capital market favors this structure because it is not a single-point narrative but offers short-term financial validation, mid-term business transformation, and long-term asset revaluation.

In the short term, AI cloud and AI revenue accounting for over half serve as catalysts for sentiment recovery. In the mid-term, AI applications and marketing determine whether Baidu can reconstruct search advertising. In the long term, Robotaxi decides whether Baidu can evolve from an internet platform into an AI infrastructure and physical-world operation platform.

Stacked together, these three layers make it inappropriate to simply label Baidu as a 'search advertising company' anymore.

III. A Turning Point for Chinese Internet: AI Moves from Press Conferences to Revenue Statements

I have always believed that the true watershed in China's internet AI competition lies not in model launches.

Press conferences can be lively, with impressive parameters and demos, but the capital market will ultimately return to three questions: Is there revenue? Is there profit elasticity? Has the valuation anchor changed?

Baidu's strength this time is that it has integrated AI into its earnings report.

This provides a new reference point for Chinese internet companies. Going forward, the market will not just evaluate Alibaba based on e-commerce GMV but also on whether AI cloud and AI-driven e-commerce can transform ad revenue. Tencent will not be judged solely on gaming and WeChat ecosystems but also on whether AI can penetrate advertising, content, office, and enterprise services. JD.com will not be viewed merely through retail profit margins but also on whether AI can enhance supply chain and hardware entry points.

AI revaluation of Chinese internet stocks cannot rely solely on thematic hype; it must ultimately be validated by performance.

Baidu now has the opportunity to become the first clear example in this revaluation. It was previously constrained by the valuation framework of its legacy business, with market concerns about slowing ad growth, weakening search entry points, and insufficient profit elasticity. However, this earnings report offers a new interpretation: Baidu's growth engine is shifting, and AI is beginning to dominate the revenue structure.

Of course, challenges remain.

The sustainability of AI cloud's high growth depends on enterprise customer demand, computing costs, and order visibility. The scalability of AI-native marketing hinges on whether advertiser budgets can migrate to new search and intelligent agent scenarios. Whether Robotaxi can become a more significant commercial asset depends on urban expansion, operational costs, and regulatory pace.

However, these uncertainties do not detract from a broader trend: Baidu has moved from left-side investments to the cusp of right-side realization.

The capital market Most afraid (most fears) stories that remain perpetually distant. Baidu's distinction this time is that AI revenue accounts for over half, intelligent cloud infrastructure revenue has surged, GPU cloud revenue has grown substantially, and Apollo Go continues to scale operations. It offers the market not a single concept but a set of financial and operational metrics to track.

In my view, this is the true value of Baidu's Q1 earnings report.

It is not merely announcing that 'Baidu is also doing AI' but telling the market that AI is no longer an Bonus question (extra credit question) for Baidu but a core question.

If the past decade of Chinese internet was dominated by traffic monetization, the next few years will likely revolve around AI realization. Only those who can integrate AI into their core businesses, write it into their revenue statements, and improve cash flow and profit elasticity will deserve valuation switching.

This is where Baidu's significance lies this time.

It is not telling a distant AI story but using its earnings report to signal that a new growth anchor has emerged for Baidu after search advertising. For Chinese internet companies, this may also be a broader signal—the AI-era competition has finally moved from press conferences to revenue statements.